CereProc has been working in speech synthesis and modification for many years and we are pleased to see this has started to become a more mainstream interest. Adobe, with their demo of VoCo highlighted the power and potential dangers of using speech synthesis to edit speech. Our CSO, Dr Matthew Aylett, was recently interviewed by Radiolab on how this technology could be used to create fake news. Listen to this thought provoking podcast here and see a bit of visual and audio modified Barack Obama here:

Colonel Sanders is now cheerfully greeting customers and taking orders at a KFC drive-thru – albeit in a robot format. KFC released the robot ‘chicken expert’ dubbed H.A.R.L.A.N.D. (Human Assisted Robotic Linguistic Animatronic Networked Device) to celebrate the National Fried Chicken Day on July 6th.

Conall Major (12) of Edinburgh visited CereProc office last week to learn all about speech synthesis. He worked with CereProc’s CSO Dr Matthew Aylett and had some fun playing around with our novelty Donald and Hillary voices. Conall said that the most difficult part of the work experience was finding something funny for the Hillary voice to say!

We think he did a brilliant job and wanted to share some of the samples Conall created.

Listen to Donald discuss his views on Mexicans and other matters:

CereProc, in partnership with the Swedish Agency for Accessible Media (MTM) , is proud to announce the release of the new text to speech voice (‘Ylva’) that will help thousands of users across Sweden.

As well as leading CereProc’s research and development team, Matthew is also a Royal Society Research Fellow and carries out research projects at The University of Edinburgh.

His impressive and in many ways unique career path is featured in the case study recently published by The Royal Society.

A team from CereProc supported by Dr Eduardo Manuel De Brito Lima Ferreira Coutinho from the Imperial College demonstrated various aspects of speech synthesis at the Science Museum in London, as part of the Royal Society’s the Next Big Thing project.

Scotland, the country best known for the scenic Highlands, whisky and deep-fried Mars bars… and impossible to pronounce place names.

Have a go first and then listen to CereProc’s Heather to see what it’s meant to sound like.

If you get all of them right, you can call yourself an honorary Scot!

Ok, let’s start easy! Can you say Edinburgh correctly? Not challenging for Scots but many a tourist (and American TV presenter) has gotten it wrong.

Mass communication service provider Alert Cascade integrated CereProc’s voices at the end of 2016 and has gone from strength to strength since then.

Lisa Leoni discusses in her latest blog post how text to speech helps her work more efficiently and improve her writing. For her new novel set in Scotland, CereProc’s Stuart is the best voice for the job.

“I discovered, and subsequently fell in love with, Scottish Stuart on Cereporc.  I’m writing contemporary romance set in Scotland with a Scottish hero and I wanted a level of authenticity in my text-to-speech experiences. *cough* Or something like that. After purchasing the voice it was super easy to install it to my machine.”

The way we interact with technology is changing - through mobile devices, Siri -like personal assistants, robotics, gaming, wearable technology and more - and synthetic speech is an integral part of that. Text to speech voices need to not only be good enough to deliver a seamless communication experience through conversational interfaces but also delight the user!